Industrial dust explosions kill and severely injure workers as well as destroy property. While the hazards of gaseous explosions are well recognized in industry, the same degree of care has not been exercised with regard to combustible dust. The agricultural industry has been especially plagued by the problem. While considerable testing relative to the explosibility of dust was done some decades ago by the Bureau of Mines, the data is now known to be apparatus dependent; it does not represent the fundamental physical quantities, and none of the data is for the layered combustible dust case. The purpose of this research program has been, and is, to determine the fundamental data, under well controlled laboratory conditions. The fundamental data consists of the turbulent burning velocity and ignition energy for a suspended dust mixture and the flame acceleration and associated overpressures for the case of dust layers in confined situations, as well as the minimum thickness of a dust layer capable of supporting combustion. This reliable data can then be applied to the myriad of situations and circumstances that arise in industry. Towards this end, two unique facilities have been developed. The first facility is a premixed turbulent combustion bomb, basically a 1.2 m diameter sphere, in which a spatially and temporally uniform turbulent dust cloud of known concentration and turbulence intensity is established. The cloud is ignited at the center and the burning velocity measured. In the second facility a uniform dust layer is spread along the bottom of a long tube. Combustion is initiated at one end and the propagation of the flame, with the accompanying pressure and temperature, is monitored as it moves downstream. In both facilities runs are made using different kinds of dust, sizes of dust, moisture content, concentration, and humidity. Much fundamental data, heretofore unavailable, including the minimum layer thickness for flame propagation through a confining structure has already been collected. The three-year research program proposed seeks to broaden the range of variables, apply more detailed instrumentation, test other dusts, determine ignition energies, determine the influence of obstacles on flame acceleration, and other related facets. From the experiments and closely coupled theoretical analysis, it is expected that a sound understanding of laminar, turbulent and layered dust flames along with flame acceleration under confined conditions (with the attendant overpressures) will be gained. These results should be useful in establishing standards for the prevention of dust explosions, developing strategies for the alleviation of dust explosions, and predicting the damage to be expected by dust explosions.